iOS 26 Brings Major Improvements to Image Playground

[Plutonius]

It's still a toy.

 

[Bustycat]

Only the anime-style results aren't scary.

 

[Bustycat]

The style of the imagery looks dated, not modern as you might think it would in a 2026 OS.
(in other words, it looks like imagery that we’ve all seen thousands of times before… so why build this capacity into the operating system?)

In fact the style from ChatGPT is actually the oldest.

 

[monkey6]

This is just so sad. Depressing that Apple thinks this is something worth promoting.

 

[ronno]

In fact the style from ChatGPT is actually the oldest.

Both look lame to me.

 

[DunedinD]

Between this and glass, Game Center’s old felt background looks tasteful now, huh.

 

[Bustycat]

Both look lame to me.

Then what's modern and not lame to you?

 

[DaveWotzit]

Nobody cares! Why not stop deflecting and develop something useful

 

[tadasZ]

just tried image playground on iPadOS 26, what a waste of time, maybe if i was 10 it would be ok, but atm complete waste of time and effort. after trying i'm feeling bad for the electricity i used while generating those peace of sh*t images.

 

[HappyMBAowner]

This is the most stupid and useless Apple feature ever done.

 

[t0rqx]

The real question: What happens with the data?

 

[klasma]

I have slight speech defect and can’t pronounce “R”. Could never speak to Siri because this dumb voice assistant doesn’t even recognise when I say “Hey Siri”…

“Hey Si-i” does work for me. Of course, Siri still has only very limited uses.

 

[MacFarmer]

file is under fart app category

 

[Jay-Jacob]

I wonder why they not added Memoji to playground? If I wanted make myself in new different ones that not included in current Memoji then could use playground but since they use photo instead of Memoji, I prefer myself in Memoji than photo version playground uses.

 

[Patchwork]

I think Image Playground only exists for Apple to say they’ve added some AI features to iOS. While some may find it useful, I deleted the app after playing with it for about half an hour. It really does remind me of iTunes Ping - nice idea, terrible implementation and late to market.

 

[svish]

Much better than what is there on iOS 18 today and the ChatGPT integration for image generation will help. Will try it out once I update my phone.

 

[Dark-Signature]

I have to say that I don't know what to do with such a solution. To be honest, I feel rather ashamed of the result that Image Playground offers. For me, it's currently just a waste of development resources!

 

[neptuna]

I'm looking forward to the update. I hate that it's so difficult to control the emotional expression of the generated image. I don't want every single image that I generate to be smiling. Especially since the teeth on these things are horrible

 

[Innovati0n]

It is much better in my testing. Beta 2 improved on Beta 1 as well.

 

[Innovati0n]

No one cares!

You don't care. That isn't the same as "no one". Main character syndrome strikes again.

 

[Apple2501]

It’s ok. You can hate me. Google is winning in usability. This image creation is centuries behind what’s available for zero dollars online. Apple still has great security and aesthetics, but Google and other ai is just more free (non restricted) and useful.

 

[ceriess]

if you ask any AI to create an image of a man with a pencil mustache, it will return an image of a man with a handlebar mustache. If you ask it to create an image based on a person wearing round glasses, the generated image will be wearing glasses of a different shape, never round. We have tried this with all commercially available AI image generators.

 

[Octavius8]

Soooooooooooooo useless

 

[routine_analyst]

So stupid. Just give me a subscription model to Midjourney, Dall-e, Gemini, Grok, ChatGPT or something else if I'm interested in prompting images into existence. Apple's going to be playing a fierce game of catch up. The world does not need another generative AI in more data centers burning through more electricity.

These other services are leaps and bounds ahead of these crummy cartoon images. Looking at any videos generated from VEO3 and comparing it to this makes it really a joke.

 

[Hallouminati]

These are still incredibly poor, particularly the top one. It is barely recognisable from the original and if there was not a point of reference I'd have no idea at all who it was.